Significant difference comparing alternatives

[Scott Miller]

How accurate is wse computation in a 2D flow area? I have several alternative scenarios modeled using full momentum, Courant number <= 1, and theta = 0.6. How different do the computed elevations need to be before we know they are not the same elevation? In this case I am not concerned about error caused by the hydrology, Manning's numbers, loss coefficients, etc. They are the same for all scenarios. I'd like to be able to say how good the computation is.

[Anonymous]

This is a very good question or just statement; heres my thoughts on it.

This depends on if you are using MAX or a time step to view depths. There is 1 depth per cell face that is calculated and shown. However, under MAX conditions this is not the case. In many cases i find the peak stage time via a hydrograph profile line or dummy weir, then i use that time series to find the time step for which the max stage occured across that line. I then utilize that depth for output. In many conditions i will create my own output shapefile consisting of merging multiple time series for the purpose of 1 output option (depending on the purpose of my study)

This of course is ALL dependent on your cell size/spacing. A WSEL is only as accurate as the terrain below it.

I hope this is what you were inquiring about

[Scott Miller]

Thank you, Luis. It sounds like you were interested in WSELs at a timestep during the maximum elevation at a particular mesh cell. MAX is never going to be a single time step if the water flows at all. I am interested in determining the confidence with which I can know the WSELs at any particular mesh cell.

Let’s say two scenarios show a difference in elevation of 0.5 foot at a given mesh cell. I could go with that. But if the computation is only accurate to 0.05 foot for either scenario, then I only know that the difference in elevation is between 0.4 foot and 0.6 foot.

I am sure computation accuracy is going to be model specific. As long as everything else is held equal, model results for a variety computation point spacing and configurations could be fit to a distribution. Distributions could be developed for mesh cells in a variety of landscape conditions. Confidence intervals could be established for each. That kind of research is beyond the scope of this project.

I was asking about something more like a rule of thumb. For instance, the older lidar data I’m using has an error of about 0.1 foot in an open, flat field. That is the best precision I have when comparing different mesh cells (not accounting for error averaging/cancelling out).

[Anonymous]

– deleted –

[cameron]

The accuracy is something you set in the control parameters. You specify a tolerance for the wse that it has to meet for it to stop iterating and move on to the next time step. If it can’t meet the tolerance you set in the number of iterations you specify, it spits out the error for that time step which would be the accuracy. This does not account for the hydrology or terrain accuracy. Garbage in equals garbage out.

[Scott Miller]

Thank you, Cameron. I am asking about confidence in that accuracy.

Installing a channel spanning box culvert where there is currently a constrictive squash pipe raises the 100-year flood elevation on a couple of residential properties by 0.02 foot. Regardless of diminutive height of this rise and immense benefits elsewhere, the regulatory requirement is zero rise.

We can and will come up with ways to abate that rise. But, really, could it be argued that there is no significant difference?

-update-

Of course, variations in mesh cell size should not yield different results for an accurate model.

[Anonymous]

You can keep reducing the cell size and the time step to the point that the model results do not change and you can make the user tolerances tighter. But even with all that, it is pretty hard to say for sure that numeric model is accurate to “x” amount.

As far as 0.02 feet for a 100-year flood:

You don’t usually know exactly what the 100-flow is going to be. You don’t have a perfect terrain model. Unless you have good data from an actual 100+year flood, you don’t actually know the Manning’s N values. Large floods frequently scour the channel bottom changing the shape of the river. You don’t usually have a perfect downstream boundary.

If the 100-flow has much velocity, you can have standing waves and debris that make the concept of 0.01 or 0.02 pretty meaningless.

And all that is without taking into account the limitations of the 2D computational program.

I think it would have been better if this country had decided to report stages to the nearest tenth of a foot, not the nearest hundredth. Unfortunately, too late for that.

But the short answer, I would say 0.02 is not significant for a 100-year flow. Now whether you could get anywhere arguing that, is a lot different question.

[cameron]

The issue with allowing one person to get away with 0.02 ft is probably not a big deal, but if multiple people do studies in the same system allowing 0.02 ft then eventually you would get above the 0.1 or 0.5 ft and we would never know as everyone says it is negligible.

The idea of comparing existing to proposed is supposed to wash out any error in accuracy as equal error would be occurring to both simulations.

One thing that is important to note is the version of HEC-RAS you are using. If you took a model from version 4.1 and ran it in version 5.0.5, the results at bridges would be quite different.

[Author]

Posts